CN213482904U - Image analysis system and device suitable for track detection - Google Patents
Image analysis system and device suitable for track detection Download PDFInfo
- Publication number
- CN213482904U CN213482904U CN202022219767.7U CN202022219767U CN213482904U CN 213482904 U CN213482904 U CN 213482904U CN 202022219767 U CN202022219767 U CN 202022219767U CN 213482904 U CN213482904 U CN 213482904U
- Authority
- CN
- China
- Prior art keywords
- track
- image
- curvature
- point
- coordinate system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The utility model relates to an image analysis system and device suitable for track detects, the system includes: the image acquisition unit is arranged on the rail transit vehicle and is used for acquiring the image of the rail when the rail transit vehicle passes through the rail; and a processing unit that processes the acquired image to determine a state of the track, the processing unit including: the device comprises an image extraction module, an image positioning module, a geometric judgment module and a track analysis module. According to the utility model discloses an image analysis system can reduce the track and detect the cost, improves track detection efficiency, has eliminated because the huge influence that motion such as train turn, ups and downs and nod produced image method track detection to can utilize the unit image to carry out three-dimensional space coordinate location to detecting the track.
Description
Technical Field
The utility model relates to an image analysis system and device suitable for track detects generally.
Background
The track position can change slowly after the railway runs for a long time. When the change exceeds its critical point, the change in the spatial position of the track accelerates. If the passenger can be corrected in time without adopting a proper detection means, the riding comfort of passengers is influenced, and the driving safety of a train can be damaged in serious cases. However, the traditional track detection mostly depends on manual work or track detection vehicles, and the detection cost is higher. Moreover, the traditional manual track detection mode is usually once every two or three months, the efficiency is low, and great risks exist if the detection is not timely. Therefore, how to quickly detect the geometric state of the existing railway line so as to timely perform necessary modification on the railway line is the most important topic for railway maintenance.
The image processing technology provides a rail diagnosis tool, which can carry out more intensive rail inspection at the colleagues of train operation and can screen dangerous road sections for further on-site inspection and maintenance. The detection means based on image processing is rapidly developed in the fields of machinery and the like, and at present, image processing means aiming at bridge displacement detection is also provided. However, the conventional image processing means has great limitations on the accuracy of the imaging lens and the observation angle of view with the object to be measured, and thus is difficult to apply to dynamic displacement detection for processing long-distance tracks. However, the motions of train turning, sinking and floating, nodding and the like have great influence on the track detection by the image method, and the single-machine image cannot carry out three-dimensional space coordinate positioning on the detected track, so that the accuracy and precision of the track detection result are greatly influenced.
SUMMERY OF THE UTILITY MODEL
In view of the above problems in the prior art, the present invention has been made to solve all or at least one of the above problems.
According to the utility model discloses an on the other hand provides an image analysis system suitable for track detects, the system includes: the image acquisition unit is arranged on the rail transit vehicle and is used for acquiring the image of the rail when the rail transit vehicle passes through the rail; and a processing unit that processes the acquired image to determine a state of the track, the processing unit including: an image extraction module configured to select an image including a straight track segment as an image to be subjected to image analysis to be started, from images including a track through which the rail transit vehicle passes; an image positioning module configured to select a linear track segment in the selected image and then set a track coordinate system and a vehicle coordinate system in the image including the linear track segment, the track coordinate system and the vehicle coordinate system being parallel, wherein the track coordinate system is: the steel rail corresponding to the linear track section is an x axis, the sleeper is a y axis, and the z axis is vertical to the x axis and the y axis; a geometric decision module configured to select an arbitrary point M on a path of the track in the track coordinate system, a curvature of the point calculated by a secant standoff method being referred to as an apparent curvature; selecting a point N, and calculating the curvature of the current moment by adopting a two-point observation method through the point M and the point N; and a track analysis module configured to determine a state of the track according to a difference between the calculated apparent curvature and the calculated curvature of the current time.
Preferably, at the geometry decision module, the apparent curvature κ of the track's course in the track coordinate system is calculated by:
wherein R is the radius of the path of the track, and the coordinate of the M point is (x)m,ym) And the coordinates of the point B are (0, y)b)，
And calculating the curvature k of the track at the current time by:
wherein R is the radius of the path of the track, and the coordinate of the M point is (x)m,ym) The coordinate of the N point is (x)n,yn)。
Preferably, the determining of the state of the track at the track analysis module includes: judging the state of the track according to a wheel-track tightness A representing relative motion between the vehicle and the track, wherein the greater the wheel-track tightness A is, the greater the risk of the vehicle derailing is represented:
wherein T is curvature estimation time corresponding to the preset running length of the vehicle, W is standard curvature, and x is standard curvaturemCorresponding curvature of M point at a predetermined value, k1And kappa2Respectively the curvature of the track curve in the track coordinate system and in the vehicle coordinate system.
Preferably, the image analysis system further comprises a calibration module configured to perform calibration of the two coordinate systems by intercepting a straight line segment between two parallel tracks in the track coordinate system and inputting the actual track gauge thereof.
Preferably, the image analysis system further comprises: and the output unit is used for inputting the image acquired by the image acquisition unit into the processing unit.
According to the utility model discloses a still another aspect provides an image analysis device suitable for track detects, the device includes: an image extraction module configured to select an image including a straight track segment as an image to be subjected to image analysis to be started, from images including a track through which a rail transit vehicle passes; an image positioning module configured to select a linear track segment in the selected image and then set a track coordinate system and a vehicle coordinate system in the image including the linear track segment, the track coordinate system and the vehicle coordinate system being parallel, wherein the track coordinate system is: the steel rail corresponding to the linear track section is an x axis, the sleeper is a y axis, and the z axis is vertical to the x axis and the y axis; a geometric decision module configured to select an arbitrary point M on a path of the track in the track coordinate system, a curvature of the point calculated by a secant standoff method being referred to as an apparent curvature; selecting a point N, and calculating the curvature of the current moment by adopting a two-point observation method through the point M and the point N; and a track analysis module configured to determine a state of the track according to a difference between the calculated apparent curvature and the calculated curvature of the current time.
The utility model provides an effectual image processing system and device carry out the analysis to the image in the train vehicle event data recorder to the orbital dynamic response of short-term test operation.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments described in the present application, and are not limited by those skilled in the art or ordinary skill.
Fig. 1 illustrates a structure diagram of an example of the middle rail detection image analysis system of the present invention.
Fig. 2 illustrates a block diagram of a processing unit in the present invention.
Fig. 3 illustrates a flowchart of the image analysis method for middle rail detection according to the present invention.
Fig. 4 illustrates the curve and secant standoff method for extracting geometric features of a middle rail according to the present invention.
Fig. 5 illustrates the dual observation method for analyzing the state of the middle rail of the present invention.
Detailed Description
Embodiments of the present invention will be described in detail hereinafter with reference to the accompanying drawings. It should be understood that the following embodiments are not intended to limit the present invention, and not necessarily all combinations of aspects described according to the following embodiments are required as to means for solving the problems according to the present invention. For the sake of simplicity, the same reference numerals or signs are used for the same structural parts or steps, and the description thereof is omitted.
Track detection image analysis system
Appendage 1 illustrates an example of the track inspection image analysis system of the present invention. Track detection image analysis system includes: the device comprises an image acquisition unit, an output unit and a processing unit.
The image acquisition unit is arranged on a rail transit vehicle to acquire the images of the rail when the vehicle runs. For example, the rail transit vehicle is a train, the rail is a train rail, and the image capturing unit may be a driving recorder 101 disposed on the train.
And the output unit is used for inputting the image acquired by the image acquisition unit into the processing unit. An example of an output unit is a video capture card 102, which transmits images captured by the tachograph to a processing unit. Alternatively, the output unit may be integrally formed with the processing unit or the image capturing unit without providing an output unit separate from the processing unit/image capturing unit.
And the processing unit is used for processing the acquired image so as to judge the state of the track. Examples of a tachograph include a PC 103, a microprocessor, a tablet, a cloud processor, and so forth. In the present embodiment, the processing unit is a PC as an example.
The tachograph 101 is disposed on the train and is used for acquiring images of the front of the train in real time during the running process of the train. The images captured by the tachograph are transmitted to the video capture card 102 and transmitted via the video capture card 102 to the PC 103 for processing and analysis.
As shown in fig. 2, the processing unit 103 may include one or more of the following modules: an image extraction module 310, a calibration module 320, an image localization module 330, a geometry decision module 340, and an orbit analysis module 350. The processing performed by each module is specifically described below in conjunction with a method for analyzing a trajectory detection image.
Track detection image analysis method
An example of the image analysis method of the present invention can be described with reference to fig. 3.
Step S100: and collecting the video signal of the track when the rail transit vehicle passes through the track, and converting the video signal into image data for storage.
For example, the drive recorder 101 collects a video signal when a train passes through a rail, and converts the video signal into an image to store the image.
Step S120: from the images including the track through which the rail transit vehicle is to pass, an image including a straight track segment is selected as an image to be subjected to the imagery analysis.
For example, among the images in the video signal acquired by the tachograph 101, an image including a straight track segment is selected for subsequent image analysis.
Step S140: selecting a straight-line track segment in the selected image, and then arranging a track coordinate system and a vehicle coordinate system in the image comprising the straight-line track segment, wherein the track coordinate system is parallel to the vehicle coordinate system: the steel rail corresponding to the linear track section is an x axis, the sleeper is a y axis, and the z axis is perpendicular to the x axis and the y axis.
For example, the plane on which the steel rail of the train is located is selected as the plane formed by the z-axis and the y-axis of the track coordinate system, and the plane on which the tachograph 101 is located in the horizontal direction is selected as the vehicle coordinate system. The two coordinate systems are parallel and spaced apart by a distance.
Alternatively, after step S140, a calibration step may be performed to calibrate the two coordinate systems by cutting a straight line segment between two parallel tracks in the track coordinate system and inputting the actual track gauge thereof.
Step S160: and a geometric judgment step of selecting an arbitrary point M on the track path in the track coordinate system, and calculating the curvature of the point by adopting a secant offset method, wherein the curvature is called apparent curvature, namely dynamic real-time curvature. And selecting a point N, and calculating the curvature of the current moment by adopting a two-point observation method through the point M and the point N.
The process of making the geometric decision is described in detail below with reference to fig. 4.
As shown in fig. 4, the determination of the track geometry is performed by the curve and secant standoff method. When there is no relative motion between the track coordinate system and the vehicle coordinate system, curve 1 of fig. 4 is the track direction, and straight line 2 is the wheel axle direction, where point B represents the intersection of the track line and the y-axis, and its coordinates are (0, y)b). c represents the distance from the intersection of the line of the track and the x-axis to the origin. From the Pythagorean theorem, the following equations 1 and 2 can be derived:
where c is the coordinate value on the x-axis of the intersection of the curve and the x-axis. R is the radius of the curve. Subtracting from the above equation yields the following equation 3:
this results in the curvature of the curve expressed by equation 4 below:
when having sideslip, float and sink, survey and roll, shake head and some when moving relatively between vehicle and the track, the rail that observes from the carriage will be followed and removed or rotatory, as shown in fig. 5, observation point M point position and the position of original observation have the difference, and the rail alignment on the image also can be distorted, the curve that sees on the image the utility model discloses in call as the sight curve, the sight curve not only the position can change, and the camber also can become, also is the camber that measures according to the above formula and measures for the sight curvature, the sight curvature is not actual camber. Both the apparent curve and the apparent curvature present a great challenge to the measurement of how to eliminate the vehicle's hunting. In order to eliminate the problem, the utility model provides a scheme of two observation points.
By adopting the method of double observation points, the relative displacement caused by the swing of the vehicle can be eliminated, and the degree of consistency of the motion of the vehicle and the track can be calculated.
FIG. 5 shows a schematic view of a double observation, where point N is another observation point, in contrast to the above, i.e.
The relationship between the distance and the curvature of the two observation points can be obtained as the following formula 5:
when the curvatures obtained by the two methods are equal, there is no relative movement between the vehicle and the track. If not equal, relative motion is generated between the vehicle and the track, and the relative motion value is the difference between the two.
Step S180: the state of the track is judged according to the curvature of the track route in the track coordinate system and the curvature of the track route in the vehicle coordinate system.
Specifically, the difference between the vehicle coordinate system and the track coordinate system is defined as the wheel-rail contact ratio Λ represented by the following formula 6, and the greater the wheel-rail contact ratio Λ, the greater the risk of derailment of the vehicle:
wherein T is a curvature estimation time, i.e., a time required for estimating a curvature, for example, a time corresponding to a predetermined length (e.g., a track length of 200m) of a vehicle running is used as the curvature estimation time, and W is a standard curvaturemThe corresponding curvature of the point M at a predetermined value (e.g., 4cm) is set as the standard curvature. Kappa1And kappa2Respectively the curvature of the course of the track in the track coordinate system and in the vehicle coordinate system.
The wheel-rail tightness is used as an index to measure the relative motion between the vehicle and the rail, and if the value is large, the risk of derailment of the train is indicated, so that the state of the rail is judged. The curvature curve characteristic is equal to the course eccentricity value when the car is not swinging relative to the track. The method thus represents the curvature of the curve with the offset values outside the secant.
The utility model discloses an image analysis adopts curve and secant standoff distance method to the secant of arc curve is as the benchmark of track coordinate, and the both ends of secant are the position of train bogie. The track coordinate system based on the secant line is easy to define even if the track has superelevation or the secant line is between the curve and the straight line. Because the track is relative track change (such as a track detection vehicle) based on the vehicle, and is not relative track change relative to geodetic coordinates, the secant coordinate system can clearly illustrate relative displacement between the vehicle and the track, and the operation of defining a plurality of track coordinate systems can be avoided.
The utility model discloses an among the image analysis system, accessible definition track, vehicle and visual angle coordinate system can avoid simultaneously because the analysis error that different image visual angles brought.
The utility model discloses an image analysis method includes the analysis of reading of image, train and track and coordinate positioning, track geometry in the image and track state. The determination of the track geometry comprises the determination of the displacement of a specific point or line on the track, and also comprises the determination of the geometric shape of a structure with parallel driving directions, and the determination of the changes of the image positions and the numerical differences of two groups of photos at different times.
Additionally, the utility model discloses an image analysis measures rail relative displacement with dynamic image measuring's mode, measures the relative dynamic displacement under different operation vehicle operation between the rail on the spot, provides the track irregularity along the line. The utility model discloses an image analysis software can carry out the track environment simultaneously, including conventional inspections such as damaged, fastener are lost, ponding to and the analysis aassessment that the vehicle is unusual or derail.
Track detection image analysis device
The utility model discloses a track detects image analysis device the device includes: the device comprises an image extraction module, an image positioning module, a geometric judgment module and a track analysis module. The structure of each module is similar to that of each module corresponding to the processing unit in the image analysis device, and is not described herein again.
According to the utility model discloses a track detection image analysis can reduce the track and detect the cost, improves track detection efficiency, has eliminated because the huge influence that motion such as train turn, ups and downs and nod produced image method track detection to can utilize the unit image to carry out three-dimensional space coordinate location to detecting the track. Compared with the prior art, the method has outstanding technical effects.
Claims (6)
1. An image analysis system suitable for track inspection, the image analysis system suitable for track inspection comprising:
the image acquisition unit is arranged on the rail transit vehicle and is used for acquiring the image of the rail when the rail transit vehicle passes through the rail; and
a processing unit that processes the acquired image to determine a state of the track, the processing unit comprising:
an image extraction module configured to select an image including a straight track segment as an image to be subjected to image analysis to be started, from images including a track through which the rail transit vehicle passes;
an image positioning module configured to select a linear track segment in the selected image and then set a track coordinate system and a vehicle coordinate system in the image including the linear track segment, the track coordinate system and the vehicle coordinate system being parallel, wherein the track coordinate system is: the steel rail corresponding to the linear track section is an x axis, the sleeper is a y axis, and the z axis is vertical to the x axis and the y axis;
a geometric decision module configured to select an arbitrary point M on a path of the track in the track coordinate system, a curvature of the point calculated by a secant standoff method being referred to as an apparent curvature; selecting a point N, and calculating the curvature of the current moment by adopting a two-point observation method through the point M and the point N; and
a track analysis module configured to determine a state of the track according to a difference between the calculated apparent curvature and the calculated curvature of the current time.
2. The image analysis system for track detection as claimed in claim 1, wherein the geometric decision module calculates the apparent curvature κ of the track path in the track coordinate system by:
wherein R is the radius of the path of the track, and the coordinate of the M point is (x)m,ym) And the coordinates of the point B are (0, y)b)，
And calculating the curvature k of the track at the current time by:
wherein R is the radius of the path of the track, and the coordinate of the M point is (x)m,ym) The coordinate of the N point is (x)n,yn)。
3. The image analysis system suitable for track detection according to claim 1, wherein the determining the track state according to the difference between the calculated apparent curvature and the calculated curvature of the current time at the track analysis module comprises: judging the state of the track according to a wheel-track tightness A representing relative motion between the vehicle and the track, wherein the greater the wheel-track tightness A is, the greater the risk of the vehicle derailing is represented:
wherein T is curvature estimation time corresponding to the preset running length of the vehicle, W is standard curvature, and x is standard curvaturemCorresponding curvature of M point at a predetermined value, k1And kappa2Respectively the curvature of the track curve in the track coordinate system and in the vehicle coordinate system.
4. The image analysis system suitable for track detection as claimed in claim 1, further comprising:
and the calibration module is configured to perform calibration of the two coordinate systems by cutting a straight line segment between two parallel tracks in the track coordinate system and inputting the actual track gauge of the straight line segment.
5. The image analysis system suitable for track detection as claimed in claim 1, further comprising:
and the output unit is used for inputting the image acquired by the image acquisition unit into the processing unit.
6. An image analysis device suitable for track inspection, the image analysis device suitable for track inspection comprising:
an image extraction module configured to select an image including a straight track segment as an image to be subjected to image analysis to be started, from images including a track through which a rail transit vehicle passes;
an image positioning module configured to select a linear track segment in the selected image and then set a track coordinate system and a vehicle coordinate system in the image including the linear track segment, the track coordinate system and the vehicle coordinate system being parallel, wherein the track coordinate system is: the steel rail corresponding to the linear track section is an x axis, the sleeper is a y axis, and the z axis is vertical to the x axis and the y axis;
a geometric decision module configured to select an arbitrary point M on the path of the track in the track coordinate system, and calculate what is called apparent curvature of the point by using a secant standoff method; selecting a point N, and calculating the curvature of the current moment by adopting a two-point observation method through the point M and the point N; and
a track analysis module configured to determine a state of the track according to a difference between the calculated apparent curvature and the calculated curvature of the current time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022219767.7U CN213482904U (en) | 2020-09-30 | 2020-09-30 | Image analysis system and device suitable for track detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022219767.7U CN213482904U (en) | 2020-09-30 | 2020-09-30 | Image analysis system and device suitable for track detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN213482904U true CN213482904U (en) | 2021-06-18 |
Family
ID=76364376
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202022219767.7U Active CN213482904U (en) | 2020-09-30 | 2020-09-30 | Image analysis system and device suitable for track detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN213482904U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113446946A (en) * | 2021-06-24 | 2021-09-28 | 中国铁道科学研究院集团有限公司 | Dynamic compensation method and device for track geometric detection data |
-
2020
- 2020-09-30 CN CN202022219767.7U patent/CN213482904U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113446946A (en) * | 2021-06-24 | 2021-09-28 | 中国铁道科学研究院集团有限公司 | Dynamic compensation method and device for track geometric detection data |
| CN113446946B (en) * | 2021-06-24 | 2023-02-17 | 中国铁道科学研究院集团有限公司 | Dynamic compensation method and device for track geometric detection data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111587202B (en) | Rail vehicle and method for measuring a track section | |
| CN101580071B (en) | Railway locomotive and vehicle operating attitude measurement system | |
| CN206781779U (en) | Laser scanning rail gauge measuring apparatus under a kind of track checking car | |
| CN113085948B (en) | Track comprehensive detection system | |
| CN112240752B (en) | Test device and test method for vehicle-mounted inspection of dynamic envelope line of rail vehicle | |
| JP6277468B2 (en) | Vehicle power pole position inspection device | |
| CN110546456B (en) | Method and apparatus for chassis surveying | |
| CN111811428B (en) | Measuring system for railway vehicle clearance | |
| KR101590712B1 (en) | Rail car and track monitoring system using running record and the method | |
| Liu et al. | A real-time posture monitoring method for rail vehicle bodies based on machine vision | |
| CN213482904U (en) | Image analysis system and device suitable for track detection | |
| CN104541124A (en) | Support detection device using laser measurement | |
| CN111895996A (en) | High-speed track detection system and method | |
| CN206787495U (en) | Multi-angle two dimensional laser scanning gage measurement instrument under track checking car | |
| JP3486239B2 (en) | Orbital deviation measuring device and method, and curvature measuring method | |
| CN113091835A (en) | Network rail tunnel comprehensive detection device and method based on edge calculation and big data | |
| CN109238149B (en) | Vehicle body attitude detection device and contact line dynamic offset detection system | |
| CN108020162A (en) | Gauge instrument and its application method based on two dimensional laser scanning Yu triangle principle | |
| CN114332788A (en) | Image analysis system, method and device suitable for track detection | |
| CN115219251A (en) | Railway vehicle dynamic performance monitoring system based on wheel pair operation posture | |
| JP2020006845A (en) | Track curvature estimation device and method | |
| KR20220096440A (en) | Rail surface defect detection sensor system combining 2d image and 3d profiler | |
| Vinkó et al. | Feasibility of onboard smartphones for railway track geometry estimation: sensing capabilities and characterization | |
| CN209492415U (en) | Rail net detection system and detection vehicle | |
| CN109000555A (en) | Vehicle bottom compensation device, system and method for bow net on-line monitoring |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |